摘要
与常规的白细胞亚分类方法不同,提出一种新型的白细胞分类方法,该方法采用希尔伯特黄变换将细胞光学信号自适应分解,并利用支持向量机构建细胞亚分类模型进行识别分类。实验结果:白细胞亚类可比性和相关性达到了血液分析仪行业标准要求。该方法为细胞信号分析以及亚分类提供了新思路。
In order to improve the accuracy of sub-classification results,a new classification method of cell recognition is proposed.In this method,the signal is decomposed into multiple intrinsic mode functions by HHT.Then,the average intensity and spectral centroid of the functions are calculated as the eigenvector.Finally,the classification mode of leukocyte subgroup is constructed by using the support vector machine.Results：leukocyte comparability and relativity achieve the standards of blood analyzer industry.This method provides a new idea for cell signal analysis and sub classification.
作者
陈铭钧
陶凌
李富贵
刘九畅
CHEN Mingjun 1,TAO Ling 1 ,LI Fugui 1,LIU Jiuchang 2(1.College of Information and Engineering,Nanchang University,Nanchang 330031,China;2.School of Information Engineering,Nahcnang Hang Kong University,Nanchang 330063,Chin)
出处
《南昌大学学报（理科版）》
CAS
北大核心
2018年第1期72-75,82,共5页
Journal of Nanchang University(Natural Science)
基金
国家自然科学基金资助项目(61261011)
江西省研究生创新专项资金项目(YC2015-S033)
关键词
信号分析
细胞分类
希尔伯特黄变换
特征向量
支持向量机
signal analysis
cell classification
Hilbert Huang Transform
eigenvector
support vector machine
